Agents that target tumor vasculature have tremendous potential as anti- cancer therapies because they do not appear to induce drug resistance. However, these agents are cytostatic rather than cytotoxic, which poses a unique set of problems for the evaluation of drug efficacy because classical measures of tumor (response" may not be valid. It will be especially important to identify their biological mechanisms of action and to develop new methods to detect these effects in primary patient tissue specimens. To this end, we have developed a technique that allows for the detection of dying tumor endothelial cells, and our preliminary results strongly suggest that anti-angiogenic agents induce apoptosis in tumor endothelial cells. The overall goal of the present studies is to determine whether endothelial cell apoptosis is a sensitive marker of efficacy in tumors treated with anti-angiogenic therapies can distinguish patients who respond to these drugs from those who do not. To this end, we propose to: (1) Define the role of specific "survival" pathways in the maintenance of endothelial cell viability in vitro. Cells will be exposed to growth factor withdrawal or VEGF receptor antagonists, and effects on signaling pathways previously implicated in cell survival (i.e. AKT) will be evaluated. We will also isolate mRNA from these cells and analyze changes in apoptosis-associated gene expression following VEGF withdrawal. (2) Determine the role of endothelial cell apoptosis in orthotopic tumor models. Nude mice bearing human pancreatic, colon, or prostate tumors will be treated with investigational agents, and effects on tumor endothelial cell apoptosis will be measured CD31/TUNEL staining. (3) Characterize the role of endothelial cell apoptosis in the effects of anti-angiogenic therapies in patients. Levels of endothelial cell apoptosis in patients treated with anti-angiogenic agents will be correlated with tumor blood flow changes and radiographic measurements of response. These studies will allow us to rapidly determine whether tumor endothelial cell apoptosis can be used as a surrogate for clinical response in patients treated with this class of novel anti-cancer agents.